ewhat resembles a
large camel's hair pencil, and may conveniently be described as a
_brush flame_. The chief advantage of a large-sized blow-pipe is, that
with it a large brush flame may be produced, which is often invaluable.
By gradually diminishing the supply of gas and air smaller brush flames
may be produced.

The jet used to supply air to the Herapath blow-pipe is usually too
fine, and consequently does not permit the passage of sufficient air to
produce a brush flame that contains excess of oxygen, even with the aid
of a very powerful blower. My own Herapath blow-pipe only gives a
satisfactory oxidising brush flame when the jet is removed altogether
from the end of the air-tube. For producing pointed flames the finer jet
of the air-tube must be used, but when a highly oxidising flame of large
size is required it must be removed. The internal diameter of the
central air-tube should be nearly half as great as that of the outer or
gas-supply tube. Fletcher's Automaton with the large air jet gives a
very liberal supply of air, and produces an excellent oxidising brush
flame. In the case of the larger-sized Automaton a consequence of this
is, however, that when fitted with the large jet it will not give so
good a pointed flame as the Herapath, which, in its turn, gives an
inferior oxidising brush. By fitting finer jets to the air-tube of
Fletcher's apparatus pointed flames can be secured when necessary.

_The Smoky Flame._--By turning on a very free supply of gas, and only
enough air to give an outward direction to the burning gas, a smoky
flame, chiefly useful for annealing and for some simple operations on
lead glass, is produced.

The Gimmingham blow-pipe and Fletcher's combination blow-pipe, in
addition to the above flames, are also adapted to produce a non-luminous
flame, resembling that of the Bunsen gas-burner, which is very
convenient for the preliminary heating of the glass, and also for
gradually cooling finished apparatus. It is not necessary to describe
the method of using these last-mentioned blow-pipes. With the more
complicated of them directions for its use are supplied.

Mr. Madan has suggested the use of oxygen in place of air for producing
the oxidising flame required for working lead glass, and to produce a
flame of high temperature for softening tubes of hard, or combustion,
glass. For the latter purpose the employment of oxygen may be adopted
with great advantage. For working lead glass, howe